book

Oracle PL/SQL for DBAs

by Arup Nanda, Steven Feuerstein

October 2005

Intermediate to advanced

454 pages

14h 44m

English

O'Reilly Media, Inc.

Read now

Unlock full access

1.2.1.1. Sections of the block1.2.1.2. Anonymous blocks
1.2.3.1. NULLs
1.2.4.1. String literals1.2.4.2. Numeric literals1.2.4.3. Boolean literals
1.2.6.1. Single-line comment syntax1.2.6.2. Multi-line comment syntax
1.3.1.1. Character data1.3.1.2. Numbers1.3.1.3. Dates, timestamps, and intervals1.3.1.4. Booleans1.3.1.5. Binary data1.3.1.6. ROWIDs1.3.1.7. REF CURSOR datatype1.3.1.8. Internet datatypes1.3.1.9. “Any” datatypes
1.3.2.1. Declaring a variable1.3.2.2. Declaring constants1.3.2.3. Anchored declarations
1.4.2.1. Simple CASE statement1.4.2.2. Searched CASE statement
1.6.2.1. RAISE statement1.6.2.2. Using RAISE_APPLICATION_ERROR
1.6.3.1. Built-in error functions1.6.3.2. Unhandled exceptions1.6.3.3. Propagation of an unhandled exception
1.7.2.1. Record-level operations1.7.2.2. Field-level operations
1.8.2.1. Using an associative array1.8.2.2. Using a nested table1.8.2.3. Using a VARRAY
1.9.1.1. Structure of a procedure1.9.1.2. Calling a procedure
1.9.2.1. Structure of a function1.9.2.2. Calling a function
1.9.3.1. Defining parameters1.9.3.2. Actual and formal parameters1.9.3.3. Parameter modes
1.9.4.1. Rules for building packages1.9.4.2. Rules for calling packaged elements1.9.4.3. Package data
1.10.3.1. Error handling with implicit cursors1.10.3.2. Implicit SQL cursor attributes
1.10.5.1. Limiting rows retrieved with BULK COLLECT
1.10.6.1. Declaring REF CURSOR types1.10.6.2. Declaring cursor variables1.10.6.3. Opening cursor variables1.10.6.4. Fetching from cursor variables
1.11.6.1. Syntax of the FORALL Statement1.11.6.2. FORALL Examples
1.13.1.1. Transaction participation1.13.1.2. Creating a DML trigger1.13.1.3. The WHEN clause1.13.1.4. Working with NEW and OLD pseudo -records1.13.1.5. Determining the DML action within a trigger
1.13.2.1. Creating a DDL Trigger
1.13.3.1. Creating a database event trigger
2.4.2.1. Strong vs. weak REF cursors2.4.2.2. REF cursor attributes2.4.2.3. Dynamic data access
3.3.2.1. Random partitioning (PARTITION BY ANY)3.3.2.2. Range partitioning (PARTITION BY RANGE)3.3.2.3. Order streaming (ORDER)3.3.2.4. Hash partitioning (PARTITION BY HASH)3.3.2.5. Cluster streaming (CLUSTER)
4.2.7.1. Using DES3GETKEY4.2.7.2. Using the key in encryption
4.3.3.1. Specifying the encryption type4.3.3.2. Specifying chaining4.3.3.3. Specifying padding4.3.3.4. Combining options in the typ parameter4.3.3.5. Handling and converting RAW data4.3.3.6. Specifying the encryption algorithm4.3.3.7. Putting it together
5.2.2.1. Problems with static policies5.2.2.2. Using a pragma
5.3.2.1. Shared static policy5.3.2.2. Context-sensitive policy5.3.2.3. Shared context-sensitive policy
5.4.3.1. Data dictionary view5.4.3.2. Event-based tracing
6.2.3.1. Turning off bind variable capture
6.2.4.1. Drawbacks with the default FGA approach6.2.4.2. Creating a user-defined audit facility
6.3.2.1. The ADD_POLICY procedure6.3.2.2. The DROP_POLICY procedure6.3.2.3. The DISABLE_POLICY procedure6.3.2.4. The ENABLE_POLICY procedure
6.4.2.1. The DBA_FGA_AUDIT_TRAIL view6.4.2.2. The FLASHBACK_TRANSACTION_QUERY view
6.5.1.1. The case for FGA6.5.1.2. The case for triggers
7.1.1.1. Controlling the precision7.1.1.2. Controlling the range
7.1.2.1. The RANDOM function
8.2.1.1. Running OS executables and anonymous blocks8.2.1.2. DBA_SCHEDULER_JOBS view
8.2.2.1. Enabling and disabling jobs8.2.2.2. Stopping running jobs8.2.2.3. Running a job8.2.2.4. Dropping a job
8.3.1.1. Examples of calendar strings8.3.1.2. Determining future calendar strings
8.7.1.1. DBA_SCHEDULER_JOB_LOG8.7.1.2. DBA_SCHEDULER_JOB_RUN_DETAILS8.7.1.3. Pruning the job log8.7.1.4. Log levels8.7.1.5. Setting the retention period
8.7.2.1. DBA_SCHEDULER_WINDOW_LOG8.7.2.2. DBA_SCHEDULER_WINDOW_DETAILS

Content preview from Oracle PL/SQL for DBAs

Transparent Data Encryption in Oracle Database 10g Release 2

When you store both the encryption key and the encrypted data in the database, another potential security hole opens up—if the disks containing the entire database are stolen, the data becomes immediately vulnerable. One way around this problem is to encrypt all the data elements and store the keys separately on a different location away from the disk drives where the data resides.

If your database is completely isolated, you may not feel that you need to encrypt its data. However, you may still want to protect the data in case of a disk theft. One solution would be to create a view to show the decrypted value. In this case, if the key is stored elsewhere, a physical disk theft will not make the data vulnerable. This approach works, but it requires an extensive and elaborate setup.

To address these types of situations, Oracle has introduced a new feature known as Transparent Data Encryption (TDE ), available starting with Oracle Database 10g Release 2. TDE uses a combination of two keys—one master key stored outside the database in a wallet and one key for each table. The same key is used for all rows in a table, and a unique key is generated for each table (as illustrated in Figure 4-7).